Transforming growth factor-betas (TGF-betas) are multifunctional regulatory proteins that play complex roles in carcinogenesis. The prevailing hypothesis is that TGF-betas have tumor suppressor activity early in the carcinogenic process, but that in the later stages, this tumor suppressor activity is lost and pro-metastatic activities dominate. This switch is thought to be accompanied by a decrease in responsiveness of the tumor cell to the tumor suppressive effects of TGF-beta, and an increase in secretion of the TGF-beta protein which then promotes the formation of a permissive environment for the tumor to develop in. Our research program uses genetically-engineered mouse and xenograft model systems to identify the biological and molecular mechanisms underlying this dual role of TGF-betas in breast cancer, and to use this information to develop novel therapeutic approaches. Our recent work suggests that the tumor suppressor activities of TGF-beta may depend on the ability of TGF-beta to regulate cancer stem cell dynamics, and to promote differentiation of cancer progenitor cells. Molecular changes that block differentiation, such as upregulation of the ID1 gene product, can block the tumor suppressing effects of TGF-beta and may contribute to the metastatic switch. In relation to the tumor-promoting role of TGF-beta in late-stage cancer, we have previously shown that mice engineered to express a protein TGF-beta antagonist are significantly protected against metastasis. Unexpectedly, there were essentially no adverse side-effects of TGF-beta antagonist treatment, so we are currently pushing this approach through further pre-clinical development using monoclonal anti-TGF-beta antibodies as the therapeutic agent. This work is being done in collaboration with Genzyme Corporation through a Cooperative Research and Development Agreement. Over the past year, our work in breast cancer model systems has suggested that anti- TGF-beta antibody treatment is effective at suppressing cancer metastasis through a combination of small effects on multiple cellular compartments, including both the tumor cell itself and components of the immune surveillance system. These data have important implications for clinical biomarker development and for understanding the biological basis of the selectivity of these agents in affecting the tumor and not the normal tissues.